Analysing the Impact of Heterogeneity with Greedy Resource Allocation Algorithms for Dynamic Load Balancing in Heterogeneous Distributed Computing System

Abstract

Heterogeneous Distributed systems have been an active research area in computer science for the last two decade, task allocation and load balancing have been a major issue associated with such systems. The load-balancing problem, attempts to compute the assignment with smallest possible makespan (i.e. the completion time at the maximum loaded computing node). This paper presents and discusses the dynamic load balancing problem on Heterogeneous Distributed Computing System (HDCS) and analyzes the impact of heterogeneity on computing capability of node on task allocation problem. Since the task assignment problem in NP hard, greedy heuristic algorithms are used to study the impact of heterogeneity on computing resources. The task model is presented as consistent ETC (Expected Time to Compute) matrix in four different heterogeneous computing environments to study the performance of heuristic algorithms to minimize the makespan.